Supply chain chapter 12

Tracking signal

- The ratio of a running total of forecast error to MAD that indicates when the pattern of forecast error is changing significantly - Essentially a comparison of forecast bias (sum of errors, rather than MFE) to forecast accuracy (MAD) over n periods

Forecast bias

- The tendency of a forecasting technique to continually over-predict or under-predict demand; also called mean forecast error

Casual models

- These models use independent observed data to predict demand. They concentrate on external factors that are thought to cause demand (such as the amount of household disposable income in the economy might be a good leading indicator of the sales of luxury items, such as sailboats)

What is the primary goal in designing a forecasting process?

- To generate forecast that are usable, timely, & accurate

What 5 steps can help managers generate forecast that are usable, timely, & accurate?

1. Identify the users & decision making processes that the forecast will support 2. Identify the likely sources of the best data inputs 3. Select forecasting techniques that will most likely transform data into timely, accurate forecast over the appropriate planning horizon 4. Document & apply the proposed technique to the data gathered for the appropriate business process 5. Monitor the performance of the forecasting process for continuous improvements

Managers try to manage demand by using variants of these 3 basic tactics

1. Influence the timing or quantity of demand through pricing changes, promotions, or sales incentives 2. Manage the timing of order fulfillments 3. Substitute by encouraging customers to shift their orders from one product to another, or from one provider to another

The CPFR process typically consist of 4 collaborative activities which are:

1. Market planning: partners collaboratively discuss such issues as the introduction of new products, store opening/closings, changing inventory prices, & product promotions 2. Demand & resource planning: customer demand & shipping requirements are forecasted 3. Execution: orders are placed, delivered, received, & paid for. This includes preparation of shipments & recording of sales. 4. Analysis: execution is monitored & key performance metrics are collected with the goal of identifying opportunities for future improvement

Operational inefficiencies across the supply chain caused by demand fluctuations

1. Requiring extra resources to expand & contract capacity to meet varying demand 2. Backlogging (delivering later than originally promised) certain orders to smooth out demand fluctuations 3. Customer dissatisfaction with the system's inability to meet all demands 4. Buffering the system through the use of safety stocks (excessive inventories), safety lead times (lead times with a cushion), or safety capacity (extra resources)

The following "rules" give an indication of how situational characteristics tend to affect forecast accuracy

1. Short-term forecast are usually more accurate than long-term forecast 2. Forecast of aggregated demand are usually more accurate than forecast of demand at detailed levels 3. Forecast developed using multiple information sources are usually more accurate than forecast developed from a single source

Statistical model-based forecasting techniques transform numerical data into forecast using one of these methods

1. Time series analyses 2. Casual studies 3. Simulation models 4. Artificial intelligence

Grassroot forecasting

- A technique that seeks inputs from people who are in close contact with customers & products (ex: sales reps.)

Seasonal index

- An adjustment factor applied to forecasts to account for seasonal changes or cycles in demand

Improvement initiatives

- Are aimed at changing information sharing systems, manufacturing & service processes, supply chain relationships, & even the product design itself

Mean percent error (MPE)

- Average error represented as a percentage of demand - Used for comparability of bias on different products

Time series analyses

- Extrapolate forecast from past data

Judgement-based forecasting

- Forecast built upon the estimates & opinions of people, most often experts working in that field of discussion - Useful when there is a lack of quantitative historical information, such as when a new product is about to be launched

Delphi method

- Forecast developed by asking a panel of experts to individually & repeatedly respond to a series of questions

Time series analysis models

- Forecasting models that compute forecast using historical data arranged in the order of the occurrence

Marketing research

- Forecasting technique that bases forecast on the purchasing patterns & attitudes of current or potential customers - Some tools used to evaluate purchasing patters include customer surveys, interviews, & focus groups

Historical analogy

- Forecasting technique that uses data & experience from similar products to forecast the demand for a new product - Can be useful when next gen. electronics are introduced (TV's, computers, phones)

Executive judgement

- Forecasting techniques that use input from high-level, experienced managers - Usually best used to make judgements regarding long-term sales or business patterns

Root mean squared error (RMSE)

- Gives an approximation of the forecast error standard deviation

Demand management

- Influencing the timing, pattern, & certainty of demand through demand management activities that adjust product characteristics including price, promotion, & availability - Especially important when customers' demands fluctuate in an unpredictable way

Big data

- Large amounts of data made available through sensors & interconnected systems

Casual studies

- Look for causal relationships between leading variables & forecasted variables

Exponential smoothing & moving average models are _______ & dont _______ the effects of a trend or any seasonal or cyclical variations in demand

- Reactive / Anticipate

Seasonality & cycles

- Regular patterns of repeating high & lows. It may be daily, weekly, monthly or even longer. For example restaurants experience these patterns during the day with peaks at breakfast, lunch, & dinner.

Simulation models

- Sophisticated mathematical programs that offer forecasters the ability to evaluate different business scenarios that might yield different demand outcomes - Try to represent past phenomena in mathematical relationships & then evaluate data to project future outcomes

Mean absolute deviation (MAD)

- The average size of forecast errors, irrespective to their directions. Also called mean absolute error - This is the simplest measure of forecast accuracy

Autocorrelation

- The correlation of current demand values with past demand values

Forecast error

- The difference between a forecast & the actual demand

Trend

- The general sloping tendency o;of demand, either upward or downward, in a linear or nonlinear fashion. New products in the growth phase of the life cycle typically exhibit an upward nonlinear trend.

Mean absolute percentage error (MAPE)

- The mean absolute deviation (MAD) represented as a percentage of demand - For purposes of comparability across products - Indicates how large errors are relative to the actual demand quantities

Forecast accuracy

- The measurement of how closely the forecast aligns with the observations over time

Weighted moving average

- A forecasting model that assigns a different weight to each periods demand according to its importance

Moving average

- A forecasting model that computes a forecast as the average of demands over a number of immediate past periods - This model is used when the demand pattern is relatively stable, without trend or seasonality

Regression analysis

- A mathematical approach for fitting an equation to a set of data - The most commonly used method for estimating relationships between leading indicators & demand

Collaborative planning, forecasting & replenishment (CPFR)

- A method by which supply chain partners periodically share forecasts, demand plans, & resource plans in order to reduce uncertainty & risk in meeting customer demand - The goal is to meet customer demand with minimal inventories, lead times, & transaction costs

Mean squared error (MSE)

- A more sensitive measure of forecast errors that approximates the error variance - Used to deal with the issue of sensitivity to the magnitude of the errors

Exponential smoothing

- A moving average approach that applies exponentially decreasing weights to each demand that occurred farther back in time

Shift or step change

- A one-time change in demand, usually due to some external influence on demand, such as a major product promotional campaign, or a sudden economic shock.

Smoothing coefficient

- A parameter indicating the weight given to the most recent demand

Postponable product

- A product designed so that it can be configured to its final form quickly & inexpensively once actual customer demand is known - In this operation system, only components, not finished goods, are stocked near sources of demand - Eliminates the need for large & complex forecasting systems

naive model

- A simple forecasting approach that assumes that recent history is a good predictor of the near future - Best with a stable demand pattern

Adaptive forecasting

- A technique that automatically adjust forecast model parameters in accordance with changes in the tracking signal

Artificial intelligence

- A broad term that refers to learning & decision making capability that stems from software algorithms -A "smart" computer program that "learns" from a combination of casual & simulation analyses using a wide array of data

Focused forecasting

- A combination of common sense inputs from frontline personnel & a computer simulation process

Stable pattern

- A consistent horizontal stream of demands. Mature consumer products such as shampoo or milk often exhibit this type of pattern